Building reputation systems for better ranking

How to rank web pages, scientists and online resources has recently attracted increasing attention from both physicists and computer scientists. In this paper, we study the ranking problem of rating systems where users vote objects by discrete ratings. We propose an algorithm that can simultaneously evaluate the user reputation and object quality in an iterative refinement way. According to both the artificially generated data and the real data from MovieLens and Amazon, our algorithm can considerably enhance the ranking accuracy. This work highlights the significance of reputation systems in the Internet era and points out a way to evaluate and compare the performances of different reputation systems.

💡 Research Summary

The paper addresses the fundamental problem of ranking objects—such as web pages, scientific articles, movies, or e‑commerce products—when the only available feedback consists of discrete user ratings (e.g., 1–5 stars). Traditional ranking approaches either treat all users equally or rely on simple averages of the ratings, which makes them vulnerable to noisy, malicious, or uninformed contributions. To overcome this limitation, the authors propose an iterative refinement algorithm that simultaneously estimates a “user reputation” score and an “object quality” score, allowing the system to weight each rating by the credibility of the rater.

The model defines two sets of variables: (R_i) for the reputation of user (i) and (Q_\alpha) for the intrinsic quality of object (\alpha). Starting from uniform or average‑based initial values, the algorithm proceeds in alternating steps. In the quality update, each object’s quality is computed as a weighted average of all received ratings, where the weight is the current reputation of the corresponding user:
\